You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
With CUDA 10.2 in Windows 10, I have built the Opt and Terra(LLVM 6.01) from master branch. Then the examples/imageWrapping is successfully built and It is able to run normally,
However, the time of every iteration is really long ( > 40s) with the figure shown below with this hardware config:
CPU: Intel(R) Core(TM) i7-8700 CPU @3.2GHz 3.19GHz
RAM: 16GB
GPU: Nvidia GeForce GTX 1080
However, when I replace the pre-built Terra with version (Release 1.0.0-beta1) and run the imageWrapping.exe again, it is found that the computation time reduces to (~15s) shown below
Terra seems like the key to the speed of the program. Is there any clues for the root cause of this problem? Thanks!
The text was updated successfully, but these errors were encountered:
With CUDA 10.2 in Windows 10, I have built the Opt and Terra(LLVM 6.01) from master branch. Then the examples/imageWrapping is successfully built and It is able to run normally,
However, the time of every iteration is really long ( > 40s) with the figure shown below with this hardware config:
CPU: Intel(R) Core(TM) i7-8700 CPU @3.2GHz 3.19GHz
RAM: 16GB
GPU: Nvidia GeForce GTX 1080
However, when I replace the pre-built Terra with version (Release 1.0.0-beta1) and run the imageWrapping.exe again, it is found that the computation time reduces to (~15s) shown below
Terra seems like the key to the speed of the program. Is there any clues for the root cause of this problem? Thanks!
The text was updated successfully, but these errors were encountered: